Blog - Announcing Ensemble | Tesera - Cloud Data Platform
Is Kubernetes ready to manage databases in production? Can we run any type of data workload? Though there are some native patterns that make database deployments a little easier, the current ecosystem is not yet mature enough. Introducing Ensemble, a simple and modular Kubernetes Operator to manage any type of database at scale.
Kubernetes has become a de facto standard for the deployment of containerized applications. While it seems to be a solved problem for stateless applications, it is an evolving topic for stateful workloads and databases.
Among all the strategies to deploy databases, the Operator pattern is the common practice since it can encode custom database orchestration logic beyond what Kubernetes itself provides. Currently, there are many Open Source operators available to deploy some of the most popular databases, many of them written by their own maintainers.
However, there does not exist a consistent workflow to run all the operators. Each one has their own specification, covers different parts of the database lifecycle or implements security and compliance requirements on their own way. This makes it hard both to adopt a database and use it in production on Kubernetes.
